Rh − POP Pincer Xantphos Complexes for C − S and C − H Activation. Implications for Carbothiolation Catalysis

: The neutral Rh(I) − Xantphos complex [Rh-( κ 3 - P,O,P -Xantphos)Cl] n , 4 , and cationic Rh(III) [Rh( κ 3 - P,O,P -Xantphos)(H) 2 ][BAr F4 ], 2a , and [Rh( κ 3 - P,O,P -Xantphos-3,5-C 6 H 3 (CF 3 ) 2 )(H) 2 ][BAr F4 ], 2b , are described [Ar F = 3,5-(CF 3 ) 2 C 6 H 3 ; Xantphos = 4,5-bis(diphenylphosphino)-9,9-dimethylxanthene; Xantphos-3,5-C 6 H 3 (CF 3 ) 2 = 9,9-dimethyl- xanthene-4,5-bis(bis(3,5-bis(tri ﬂ uoromethyl)phenyl)-phosphine]. A solid-state structure of 2b isolated from C 6 H 5 Cl solution shows a κ 1 -chlorobenzene adduct, [Rh( κ 3 - P,O,P -Xantphos-3,5-C 6 H 3 (CF 3 ) 2 )(H) 2 ( κ 1 -ClC 6 H 5 )][BAr F4 ], 3 . Addition of H 2 to 4 a ﬀ ords, crystallographically characterized, [Rh( κ 3 - P,O,P -Xantphos)(H) 2 Cl], 5 . Addition of diphenyl acetylene to 2a results in the formation of the C − H activated metallacyclopentadiene [Rh( -Xantphos)( σ , κ 1 -4-(COMe)C 6 H 3 SMe)(H)][BAr F4 ], 14 . The temporal evolution of carbothiolation catalysis using mer - κ 3 - 8 , and phenyl acetylene and 2-(methylthio)acetophenone substrates shows initial fast catalysis and then a considerably slower evolution of the product. We suggest that the initially formed fac -isomer of the C − S activation product is considerably more active than the mer -isomer (i.e., mer - 8 ), the latter of which is formed rapidly by isomerization, and this accounts for the observed di ﬀ erence in rates. A likely mechanism is proposed based upon these data.


INTRODUCTION
The transition-metal chemistry associated with diphosphine, and related, pincer ligands is significant to many areas of organometallic chemistry, catalysis, and materials science. This is due to the attractive properties that such ligands give the resulting metal complexes, for example: stability, electronic and steric variability, and, in some cases, metal−ligand cooperativity. 1−5 Central to many of these studies have been systems based around "PCP" and "PNP"-type ligand scaffolds ( Figure  1). These provide relatively rigid ligand environments, that often result in mer-κ 3 -coordination geometries at the metal center. A less explored class of pincer ligand comes from POPframeworks based upon the Xantphos-motif (Xantphos = 4,5bis(diphenylphosphino)-9,9-dimethylxanthene, Figure 1), 6,7 that although originally developed to be cis-chelating bidentate phosphine ligands can also act as mer-κ 3 8 or, less commonly, fac-κ 3 POP-pincer ligands. 9 Recently there has been significant interest in the chemistry associated with Xantphos acting as a POP-pincer ligand to the group 9 metals (Figure 2), for example in carbonylation (A), 9 activation of H 2 /alkenes (B− D), 10−12 silylation (E), 13 hydroamination (F), 14 hydroacylation (G), 15 and amine−borane dehydrocoupling and hydroboration processes (H). 16 −18 We have recently reported the use of the [Rh(Xantphos)] + catalyst fragment in carbothiolation reactions between alkynes and ketone-bearing aryl sulfides to produce alkenyl sulfide products (Scheme 1). 19 These catalyst systems operate best at elevated temperatures, e.g. coupling phenyl acetylene and 2-(methylthio)acetophenone at 373 K in chlorobenzene or dichloroethane solvent. In addition to this atom-efficient C− C and C−S bond formation process, the resulting alkenyl sulfide can be used for further functionalization, i.e. functional group recycling. This Xantphos system delivers improved yields and conversion times compared to our previously reported DPEphos system 20 [cf. Xantphos 99%, 2 h; DPEphos 90%, 24 h, Scheme 1, DPEphos = oxidi-2,1-phenylene-bis-(diphenylphosphine)]. For the DPEphos catalyst, in situ NMR spectroscopic studies identified the C−S activated intermediates fac-κ 3 -I and mer-κ 3 -J to be in equilibrium with one another at 298 K, while at 333 K mer-κ 3 -J was the observed resting-state during catalysis. 20 Transition-metal-catalyzed C−S activation processes are becoming increasingly important in synthesis, 21 especially their use in cross coupling reactions. 22−26 In this contribution we describe a study of the C−S activation processes occurring in the [Rh(Xantphos)][BAr F 4 ] system. We also report a new, readily accessible, Rh(I)−Xantphos precursor complex which has a labile alkyne ligand; report C−H activation processes mediated by this fragment; and also make comment upon the likely active species in carbothiolation catalysis.

Synthesis of Rh(I) Precursor Complexes.
A suitable entry point into the C−S activation chemistry would be a Rh(I) precursor that can promote oxidative addition of the ketosulfide substrate. Previously in the DPEphos system we used the orthoxylene-ligated [Rh(κ 2 -P,P -DPEphos)(o-xylene)][BAr F 4 ] as a suitable precursor. This was generated by addition of H 2 to the corresponding NBD complex in o-xylene (NBD = norbornadiene). 20 Noting that our recent report of the alkyne carbothiolation catalyzed by [Rh(Xantphos)] + used chlorobenzene as a solvent, we targeted an appropriate precursor that involved this as a ligand. Addition of H 2 to previously reported cis-[Rh(κ 2 -P,P -Xantphos)(NBD)][BAr F 4 ], 1a, 15 in C 6 H 5 Cl solvent resulted in the quantitative formation of a Rh(III) dihydride complex [Rh(κ 3 -P,O,P -Xantphos)(H) 2 ][BAr F 4 ], 2a, as characterized by in situ 1 H and 31 P{ 1 H} NMR spectroscopy (Scheme 2), which is analogous to recently reported variants Xantphos-t Bu 11 or Xantphos-i Pr 11 . Complex 2a could not be isolated as a solid. However, treatment of the 3,5-C 6 H 3 (CF 3 ) 2 P-substituted Xantphos analog of 1a (complex 1b) with H 2 in C 6 H 5 Cl solution led to the isolation of [Rh(κ 3 -P,O,P -Xantphos-3,5-C 6 H 3 (CF 3 ) 2 )(H) 2 (κ 1 -ClC 6 H 5 )][BAr F 4 ], 3, as a white crystalline solid in 79% yield. The solid-state structure of the cation as determined by single crystal X-ray diffraction is shown in Figure 3. The solid-state structure of 3 shows a cis-dihydride species (the hydrides were located in the difference map), with a κ 1chlorobenzene [Rh−Cl 2.5207(12) Å] and a mer-κ 3 -Xantphos ligand. Complex 3 is closely related to similar compounds in which acetone, MeCN, 15 or H 3 B·NMe 3 16 replace the chlorobenzene ligand, and the gross structural metrics are similar to these adducts. In the 1 H NMR spectrum of complex 3 in C 6 D 5 Cl solution, a single environment, indicated by a broad signal at δ −17.7, is observed for the hydride ligand, and a single Xantphos-methyl signal (6 H) is observed. A single environment showing coupling to 103 Rh [δ 45.3, J(RhP) = 118 Hz] is also observed in the 31 P{ 1 H} NMR spectrum. These data suggest a time-averaged C 2v symmetry for the metal fragment, likely approximated to a 5-coordinate "Y-shaped" [Rh(κ 3 -P,O,P -Xantphos-3,5-C 6 H 3 (CF 3 ) 2 )(H) 2 ][BAr F 4 ], i.e. 2b. When complex 3 is dissolved in CD 2 Cl 2 , a signal assigned to free C 6 H 5 Cl is observed, in addition to very similar NMR spectroscopic data to 2a in CH 2 Cl 2 solution. Complexes 2a/b in solution, or 3 in the solid state, do not lose H 2 on application of a vacuum (∼10 −3 Torr). Cooling of 3 (183 K, CD 2 Cl 2 ) did not result in a 1 H NMR spectrum that reflects the solid-state structure (i.e., inequivalent hydrides were not observed). This is in contrast to [Rh(κ 3 -P,O,P -Xantphos)(H) 2 (acetone)][BAr F 4 ], 15 in which the hydrides are equivalent at 298 K but inequivalent at low temperature, which is suggested to be due to reversible decoordination of acetone. Given these data, we suggest that complex 3 exists in C 6 H 5 Cl or CD 2 Cl 2 solution predominantly as [Rh(κ 3 -P,O,P -Xantphos-3,5-C 6 H 3 (CF 3 ) 2 )(H) 2 ][BAr F 4 ], 2b, but can be recrystallized from C 6 H 5 Cl solvent to give the solvent adduct 3. We discount a time-averaged six-coordinate complex with trans-hydrides as this would likely have a chemical shift for the hydride ligands at higher field than is observed [e.g., [Rh(κ 3 -P,O,P -Xantphos-t Bu)(H) 2 (OH 2 )][SbF 6 ] δ −9]. 11 Complex 3 adds to the very small number of crystallographically characterized κ 1 -chlorobenzene complexes. 27 We cannot comment on whether a similar complex would be formed with 2a, as we have been unable to produce crystalline material suitable for analysis.
As noted recently by Goldman, the formation of a Rh(III) dihydride complex over a Rh(I) species on addition of H 2 could well be due to the relatively weak trans-influence of the POP− ether ligating group in Xantphos that favors the product of oxidative addition of H 2 . 11 In addition to this, wide-bite angle ligands such as Xantphos will tend to favor Rh(III) (dihydrides) over Rh(I) species. 28 Thus, avoiding the use of H 2 in the production of the precursor for C−S activation would thus be advantageous in generating a Rh(I) precursor. To accomplish this, addition of Xantphos to [Rh(COE) 2 Cl] 2 (COE = cyclooctene) in C 6 H 6 solution resulted in the formation of an insoluble (benzene) material that we tentatively characterize as the Rh(I) complex [Rh(Xantphos)-Cl] n , 4, presumably formed as a coordination polymer with bridging halide ligands (Scheme 3), based upon its insolubility. This material analyzed correctly for composition by elemental analysis. Dissolution in CH 2 Cl 2 resulted in a mixture of, uncharacterized, products. Interestingly, the analogous com- Scheme 3. Scheme Relating the Synthesis of Complexes 4, 5, 6, and 7 a plexes with Xantphos− t Bu 11 or Xantphos− i Pr 12 are soluble in benzene and shown to be monomers in the solid state. This presumably reflects the differing steric demands of the Psubstituents. Suspension of complex 4 in toluene and addition of H 2 resulted in the formation of the neutral, soluble, Rh(III) cis-dihydride complex Rh(κ 3 -P,O,P -Xantphos)(H) 2 Cl, 5, obtained in 79% yield (Scheme 3) as colorless crystals. The solid-state structure of 5 is shown in Figure 3 and is very closely related to recently reported Rh−Xantphos− t Bu 11 and Ir− Xantphos− i Pr 12 analogues (i.e., B Figure 2). The NMR spectroscopic data for 5 are in full accord with the solid-state structure and show inequivalent hydride environments in the 1 H NMR spectrum, in contrast to complex 3. The formation of 5 in good yield from simple addition of H 2 to 4 further supports the latter's formulation.
Removal of the chloride ligand in 4 is easily achieved by treatment of a suspension in 1,2-C 6 H 4 F 2 solvent with Na[BAr F 4 ] in the presence of a slight excess of diphenylactylene to yield the Rh(I) complex mer-[Rh(κ 3 -P,O,P -Xantphos)(η 2 -PhCCPh)][BAr F 4 ], 6, as orange crystalline material in good (82%) yield on work-up. The solid-state structure of the cation in complex 6 is shown in Figure 4, and this shows the alkyne to be bound perpendicular to the Rh−POP plane (angle between planes = 90.7°). The C−C distance [1.256(6) Å] is lengthened compared to free ligand [1.206(2) Å 29 ]. The NMR spectroscopic and microanalytical data support this formulation; for example, a single environment is observed at δ 20.4 [d, J(RhP) = 124 Hz] in the 31 P{ 1 H} NMR spectrum, while one, relative integral 6 H, signal is observed for the Xantphos methyl groups. There are surprisingly few pseudo square planar complexes reported which have pincer ligands and bound alkynes or alkenes. One example is (PCP)Ir(trans-1,4-phenylbut-3-ene-1-yne) [PCP = (κ 3 -C 6 H 3 -2,6-(CH 2 P t Bu 2 ) 2 )], which also adopts a similar coordination mode with the metal−pincer fragment to 6 and shows a similar C−C distance [1.270(4) Å]. 30 Complex 6 represents an excellent starting point into the C−S activation chemistry, which is described in the next section.
An alternative methodology for generating complex 6 is addition of H 2 (1 atm) to complex 1a in CH 2 Cl 2 solution to form complex 2a, evacuation, followed by addition of a CH 2 Cl 2 solution of diphenylacetylene (Scheme 5). Under these conditions of excess alkyne, cis-stilbene is also observed to be formed by 1 H NMR spectroscopy [δ 6.57]. Also produced is a parallel product, identified by NMR spectroscopy and single crystal X-ray diffraction as [Rh(κ 3 -P,O,P -Xantphos)(ClCH 2 Cl)-(σ,σ-(C 6 H 4 )C(H)CPh)][BAr F 4 ], 7. Complex 7 is formed as a major product relative to 6 under these conditions (relative ratio 2:1 of 7 to 6). Although single crystals of complex 7 could be produced in this manner, in bulk form it was always contaminated with some 6.
The observation that complex 7 is preferentially formed over complex 6 when complex 2a is formed in situ suggests that this latter Rh(III) complex is central to the C−H activation process to form 7. Complex 6 does not undergo C−H activation to form 7 (Scheme 4), supporting this hypothesis. The C−H activation process can be reversed. Complex 7 reacts with H 2 (1 atm) slowly (12 h, CD 2 Cl 2 ) to form complex 2a as determined by 31 P{ 1 H} NMR spectroscopy. No cis-stilbene was observed to be formed [δ 6.57, CDCl 3 ] under these conditions, and when the reaction was performed with D 2 , signals at δ 7.25 and 7.57 were observed in the 2 H NMR spectrum, consistent with formation of trans-PhCHC(D)(C 6 H 4 D) [lit. 1 H δ 7.19, 7.60 CDCl 3 38 ]. d 2 -2a was also observed to be formed by a broad signal at ∼δ −18 in the 2 H NMR spectrum. This shows that under conditions exogenous of H 2 , complex 7 can be recycled back to form 2a, but only slowly.
A suggested mechanism for the formation of complex 7 is shown in Scheme 5. Addition of H 2 to 1a forms the dihydride complex 2a, which then undergoes coordination and hydride insertion with diphenylacetylene to give the corresponding cisdiphenylvinyl intermediate. The pathway then bifurcates: Elimination of cis-stilbene (observed) and subsequent coordination of diphenylacetylene eventually give 6. Alternatively a cis−trans isomerization (possibly via a metallacyclopropene intermediate) 31,39 followed by C−H activation at the orthoposition of arene occurs, likely via a sigma−CAM process (sigma−complex assisted metathesis), 40 loss of H 2 , and coordination of CH 2 Cl 2 to form 7.
2.2. C−S Activation Chemistry. With a suitable Rh(I) precursor to C−S activation in hand in complex 6, reactivity with a variety of arylsulfides was explored ( Figure 5). Given the use of the substrate I [2-(methylthio)acetophenone] in our recently reported carbothiolation chemistry, 19,20 initial studies focused upon using this starting material. Addition of I to a toluene solution of 6 resulted, at 298 K, in the formation of a mixture of two isomers, identified spectroscopically as fac-and mer-[Rh(κ 3 -P,O,P -Xantphos)(σ,κ 1 -C 6 H 4 COMe)(SMe)][BAr F 4 ], 8, Scheme 6. These isomers were initially formed in a 20:1 ratio respectively after 20 min and equilibrated over time to give κ 3mer-8 as the thermodynamic product after 32 h. This process may be accelerated by heating to 373 K for 40 min, after which time κ 3 -mer-8 can be isolated in good (70%) yield as the only product. In both cases free diphenylacetylene is also observed to be formed. The 31  . Such a pattern is characteristic of 31 P environments in which one sits opposite to a high trans influence ligand (such a SMe) while one sits opposite a weaker ligand, i.e. the ketone. 39,41 The 1 H NMR spectrum shows two signals at δ 1.56 and 1.43 (both of relative integral 3 H) as doublets [J = 7.6 and 1.2 Hz, respectively]. These are assigned to the SMe and ketone−methyl group, respectively, and the larger coupling to 31 P in the former (confirmed by selective decoupling experiments) is fully consistent with its trans disposition to a phosphine. 20,39 Although we have not obtained a solid-state structure for κ 3fac-8, similar complexes have been crystallographically characterized with DPEphos. 10,39,41 The NMR data for κ 3 -mer-8 are also fully consistent with its formulation; notably, a single environment is observed in the 31 P{ 1 H} NMR spectrum [J(RhP) = 106 Hz], and the SMe [δ 0.87] and ketone−methyl groups [δ 1.99] are observed as singlets in the 1 H NMR spectrum; consistent with their now cis-relationship with the phosphines. Two methyl environments are observed for the Xantphos ligand. The solid-state structure of κ 3 -mer-8 is shown in Figure 6. In the unit cell there are two crystallographically distinct cations, for which the structural metrics are very similar. The structure demonstrates the κ 3 -mer geometry of the final product, in which the thiomethyl and aryl groups sit mutually   Exchange between isomers of Xantphos, in which the phosphines adopt a cis or trans geometry, has been noted previously, and in some cases this is reversible. 10,42,43 For the system fac/mer-[Rh(κ 3 -P,O,P -Xantphos)(PCyp 3 )(H 2 )][BAr F 4 ], equilibrium measurements indicate that the fac isomer is preferred and that although enthalpically there is little difference between the isomers, entropic considerations determine the final position of the equilibrium (ΔS is positive for the generation of the fac isomer). 10 On heating mer-8 (373 K, C 6 D 5 CD 3 ), we saw no evidence, to the detection limit of 31 P NMR spectroscopy, for fac-8. However, we cannot discount its formation at a very low equilibrium concentration under these conditions.
Using the protocol for the isolation of pure C−S activated product developed above, the 2-arylmethylsulfides II−V were combined with complex 6 to afford complexes 9−12 in good yields (66%−78%), all as the mer-isomers, Scheme 7. These red/brown complexes have been characterized by NMR spectroscopy, ESI−MS, microanalysis, and X-ray crystallography (apart from complex 9, in which extensive disorder of the cation precluded a reliable analysis). Solid-state structures of the cations in complexes 10 and 11 are shown in Figure 7 (complex 12 is presented in the Supporting Materials) and are closely related to that of mer-8. For the structures 10−12, the activated thio−aryl ligand is equally disordered between two orientations in which the SMe and ortho-coordinating groups have swapped positions, giving the overlaid disordered structures C 2v symmetry in the solid state. Although this disorder could be modeled satisfactorily, this does mean that caution should be exercised in the interpretation of the fine details of the structural metrics around the metal. Nevertheless, the structures confirm the formulation.
When allyl phenyl sulfide VI is combined with complex 6, the C−S activated product fac-[Rh(κ 3 -P,O,P -Xantphos)(η 3 - 13, is isolated as a green crystalline solid in 60% yield, Scheme 8. The solid-state structure of 13 is shown in Figure 8 44,45 Presumably this process occurs via initial coordination of the alkene followed by oxidative cleavage of the C−S bond. 44 Very few examples of crystallographically characterized Xantphos (or DPEphos) complexes that also contain allyl ligands have been reported, and all have a κ 2 -P,P geometry of the chelate ligand, 46 different from the κ 3 -POP observed for complex 13.
The reaction of (4−methylthio)acetophenone VII with complex 6 probes the requirement for a ortho-directing group in these C−S activations, Scheme 8. The product of this reaction is mer-[Rh(κ 3 -P,O,P −Xantphos)(σ,κ 1 -C 6 H 3 (COMe)-SMe)(H)][BAr F 4 ], 14, isolated in moderate yield (38%) as colorless crystalline material, for which NMR spectroscopy and a single crystal X-ray diffraction study showed it to be the product of C−H activation ortho to the directing ketone group. The 1 H NMR data for 14 show a single hydride environment at δ −14.99 (relative integral 1 H) as a doublet of triplets and the absence of a methyl resonance below δ 1 that would be indicative of a Rh−SMe group. The 31 P{ 1 H} NMR spectrum displays a single environment showing coupling to a Rh(III) center [δ 39.5, J(RhP) = 114 Hz]. These data suggest C−H activation has occurred rather than C−S activation, and the solid-state structure derived from a single crystal X-ray analysis confirms this, Figure 8. This shows that the 4-(methylthio)acetophenone has undergone C−H activation ortho to the ketone to afford a cis-hydridoaryl motif, with the hydride ligand located in the final Fourier difference map. The rhodium's coordination sphere is completed by coordination of the ketone through a dative bond, similar to that observed for the C−S activated products. Such reactivity is directly related to the directed activation of ortho-C−H bonds first reported by Murai, 47 in which arylketones underdo hydroarylation with alkenes. 48 Complex 14 does not react further with alkynes, for example phenyl acetylene.
2.3. Alkyne Trimerization. Complex 6 does not react further with diphenylacetylene. However, it does react slowly (20 h, 298 K) with phenylacetylene to afford the products of cyclotrimerization, 1,2,3-Ph 2 C 6 H 3 and 1,2,4-Ph 3 C 6 H 3 , as identified by 1 H NMR spectroscopy and GC−MS. These are formed in a 2:1 ratio (Scheme 9). Complex mer-8 will also promote this process, but only at elevated (373 K) temperature, and there is no reaction at 298 K. This latter reaction presumably occurs via initial carbothiolation to release the alkenyl sulfide product (observed by 1 H NMR spectroscopy), and the resulting [Rh(Xantphos)] + fragment is then free to mediate cyclotrimerization. These trimerization products are also observed during carbothiolation catalysis but in low relative yield (less than 20%), meaning that although this process is competitive with carbothiolation in this system, use of a slight excess of alkyne (2 equiv) means that full conversion of the ketone is achieved.
2.4. Catalysis. Under room temperature conditions (298 K) complex 6 will only slowly catalyze the carbothiolation of I with phenyl acetylene (55% conversion at 24 h). However, heating to 373 K results in faster turnover (6 h) with full conversion to the alkenyl sulfide product. Complex mer-8 was the only observed organometallic product during this catalysis [by 31 P{ 1 H} NMR spectroscopy]. Starting from mer-8, the reaction is considerably slower (16 h, 373 K), and again, mer-8 is the only observed species during catalysis. Following these reactions by in situ NMR spectroscopy during the early stages of catalysis using 1,3-dimethoxybenzene as internal standard (Figure 9) revealed the temporal profiles for the formation of the product for both starting pre-catalysts. For catalysts using complex 6, initial very rapid catalysis (∼15% conversion after 20 min) is then followed by a much slower rate. These data are consistent with the initial formation of a very active catalyst that turns over the carbothiolation rapidly but quickly forms a less active catalyst. We suggest, based on our stoichiometric experiments, that this can be explained by the initial formation of fac-8. In principle this could decoordinate its POP ether linkage relatively easily to allow access to the alkyne to the now 16-electron metal center. Such potentially hemilabile behavior has been noted for related Xantphos complexes on reaction with MeCN. 10 Isomerization to the presumably less active mer-8, shown to be rapid at 373 K, results in a slower rate of catalysis. Consistent with this, mer-8 is the only observed organometallic species during catalysis, as the resting state, while starting from mer-8 results in no rapid initial burst of activity, but only slow turnover at a rate broadly similar to that for the slow regime when starting from 6. Whether catalysis in   slow regime occurs by a small (undetectable by 31 P NMR spectroscopy) equilibrium concentration of fac-8 or an alternative active species is not clear. A suggested pathway for catalysis is shown in Scheme 10. This has close similarities for those suggested for alkyne carbothiolation 20 and alkyne hydroacylation 39 using the [Rh(DPEphos)] + system. We have observed related intermediates previously in alkyne hydroacylation systems with [Rh(DPEphos)] + fragments. 39,41 Given the scatter present in the data (Figure 9), we are reluctant to speculate further the likely rate-determining steps or the order of reaction, and although the broad temporal profile was reproducible (e.g., mer-8 versus 6), there was significant variation between repeat runs with regard to the approximate rate in the slow regime. Tests for homogeneous/ heterogeneous behavior indicated a homogeneous catalyst 49 (Hg-test, sub-stoichiometric phosphine, microfiltration tests; a clear red solution is observed during catalysis). However, we cannot completely rule out the presence of a small amount of colloidal Rh-catalyst, or decomposition to a different active species. Notably the Hg-test still showed an initial burst of reaction rate followed by slower turnover, consistent with our mechanistic hypothesis. It is also interesting to note that when catalysis is performed in dichloroethane solvent at 353 K (rather than toluene at 373 K), a very different profile is observed characteristic of a greater than zero order process: the reactions are much faster and the reaction solutions are observed to be black. 19 Whether this indicates a change from homogeneous to heterogeneous catalysis under these different conditions is as yet unclear, but it is notable that nanoparticle formation has been shown to be strongly dependent on solvent polarity. 50,51

CONCLUSIONS
We report the synthesis of mer-[Rh(κ 3 -P,O,P -Xantphos)(η 2 -PhCCPh)][BAr F 4 ], 6, which acts as a latent source of the [Rh(κ 3 -P,O,P -Xantphos)] + fragment. That this complex comes from simple halide abstraction from [Rh(κ 3 -P,O,P -Xantphos)Cl] n suggests that this might well be a useful methodology for future synthetic and catalytic studies. Demonstrating the utility of 6 in this regard, we have shown that it readily undergoes ketonedirected C−S and C−H activation processes with aryl− ketosulfides. Additionally we have also reported rare examples of Rh−chlorobenzene and Rh−dichloromethane complexes. The temporal evolution of catalysis using 6 shows initial fast catalysis and then and then much slower turnover, and we suggest that the fac-isomer of the C−S activation product (i.e., fac-8) is considerably more active than the mer-isomer (i.e., mer-8), the latter being formed rapidly by isomerization. A future strategy for optimizing this catalyzed carbothiolation reaction might well be to limit the flexibility of the ligand set so that this isomerization does not occur, and studies are currently ongoing to study the effect of incorporating such ligands into the catalyst structure.
Synthesis of 2-Nitrothioanisole. To a stirred solution of 1fluoro-2-nitrobenzene (0.268 mg, 1.9 mmol) in DMF (10 mL) was added a sodium thiomethoxide 21% aqueous solution (2.2 mmol). After the addition was complete, the solution was stirred at room temperature for 2 days. Then 100 mL H 2 O was added to precipitate the product. The solvent was decanted off and the resulting solid was recrystallized (dichloromethane/petrol) to give the sulfide as pure product (167 mg, 52%). The 1 H NMR data is consistent with the reported values.    [Rh(Xantphos)Cl] n (4). To a J. Youngs tube containing [Rh-(COE) 2 Cl] 2 (305 mg, 0.425 mmol) and Xantphos (492 mg, 0.85 mmol) was added 10 mL benzene, sealed, and heated to 80°C for 4 h. The solution changed color to dark red, followed by product precipitation. The resulting precipitate was filtered and washed with pentane. The precipitate was dried under vacuum to yield a brick-red powder. Yield = 91%, 555 mg. Complex 4 was insoluble in benzene, acetone, and reacts with dichloromethane to give an unidentified mixture. Toluene can be replaced for benzene in the procedure with no significant change in yield.
Microanalysis mer-[Rh(κ 3 -P,O,P -Xantphos)(H) 2 Cl] (5). The title compound was formed by placing a solution of 4 (50 mg, 7.0 × 10 −2 mmol) in toluene (5 mL) under an atmosphere of H 2 (1 atm). The solution was stirred for 16 h at room temperature. The solution was degassed and was reduced to ∼1 mL, 10 mL hexane was added to the solution to precipitate the product, the solid product washed with hexane and dried in vacuo. Yield = 79%, 40 mg. The crystals suitable for X-ray diffraction were obtained by layering the benzene solution of the title compound with hexane at room temperature.  (7). 1a (246 mg, 0.15 mmol) was dissolved in dichloromethane (5 mL), and the solution was exposed to H 2 for 30 min (1 atm). Then the solution was degassed by evaporation to dryness, diphenylacetylene (107 mg, 0.60 mmol) in CH 2 Cl 2 (5 mL) added, and the mixture was stirred for 1h. The solution was evaporated and washed with hexane, The solid product was recrystallized with dichloromethane and hexane at −18°C. Total yield =70%, 180 mg. The two products 6 and [Rh(κ 3 -POP-Xantphos)(ClCH 2 Cl)(σ,σ-(C 6 H 4 )C(H)CPh)][BAr F 4 ] 7 were observed by NMR spectrocopy, the ratio being 1:2 respectively. Crystals of 7 suitable for X-ray diffraction were obtained from dichloromethane solution with pentane at −18°C.   (12). To a J. Youngs tube containing 6 (25 mg, 0.015 mmol), 2-(2-(methylthio)phenylpyridine (4 mg, 0.02 mmol) was added 10 mL chlorobenzene, sealed and the resulting solution was stirred for 3 h at 80°C. The solution was concentrated to dryness under vacuum, the resulting residue washed with pentane, then recrystallized by diffusion of pentane into a dichloromethane and toluene mixture solution of the crude product at −18°C to yield red crystals. Yield = 69%, 18 mg.  (13). To a J. Youngs tube containing 6 (100 mg, 0.06 mmol), allyl phenyl sulfide (19 mg, 0.12 mmol) was added 10 mL chlorobenzene, sealed and the resulting solution was stirred for 2 h at 80°C. The solution was concentrated to dryness under vacuum, the resulting residue washed with pentane, then recrystallized by diffusion of pentane into a dichloromethane solution of the crude product at −18°C to yield green crystals. Yield = 60%, 61 mg.   (14). To a J. Youngs tube containing 6 (80 mg, 0.046 mmol), 4-(methylthio)acetophenone (11.5 mg, 0.069 mmol) was added 10 mL toluene, the resulting solution was stirred for 2 days at room temperature. The solution was filtered, and then concentrated to dryness under vacuum, the resulting residue washed with pentane, then recrystallized by diffusion of pentane into a chlorobenzene solution of the crude product at −18°C to yield colorless crystals. Yield =38%, 30 mg. Crystallography. X-ray crystallography data were collected on an Agilent SuperNova diffractometer using graphite monochromated Cu Kα radiation (λ = 1.54180 Å), or an Enraf Nonius Kappa CCD diffractometer using Mo Kα radiation (λ = 0.71073 Å) with the use of low-temperature devices [150(2) K]. 61 Data were reduced using the instrument manufacturer software, DENZO-SMN, 62,63 and Crystal-Clear. All structures were solved ab initio using SIR92, 64 or SuperFlip, 65 and the structures were refined using CRYSTALS 66 or SHELX. 67 All non-hydrogen atoms were refined with anisotropic displacement parameters. All hydrogen atoms were placed in calculated positions using the riding model unless otherwise stated. Crystallographic data have been deposited with the Cambridge Crystallographic Data Center under CCDC 1025603−1025612. These data can be obtained free of charge from The Cambridge Crystallographic Data Center via www.ccdc.cam.ac.uk/data_request/ cif.